Phenomenological CALPHAD (CALculation of PHAse Diagrams) models, widely used for multicomponent materials, often contain a considerable number of parameters and require fitting using data from a relatively small number of experimental measurements or theoretical calculations. Sometimes these parameters are introduced for the purpose of improving model fits but without clear physical justification, which leads to overparametrized models with poor generalization performance. Automated approaches for optimal model selection based on the available data therefore become critical. In this work, a least absolute shrinkage and selection operator (LASSO)-based approach is developed for model selection by leveraging the linearity of the CALPHAD model with respect to its parameters to convert the model selection and fitting to a LASSO minimization problem. We demonstrate its utility for thermodynamic modeling of thermochemical hydrogen (TCH) production materials using lanthanum strontium manganite (LSM) as an example. Various TCH-relevant properties, including oxygen stoichiometry as a function of oxygen partial pressure, enthalpy of reduction, and entropy of reduction, are successfully predicted with reasonable accuracy using a minimal set of model parameters. Importantly, the model selection and fitting involve minimal human decision; it can therefore be applied to high-throughput DFT defect calculations and yield efficient workflows for TCH material modeling and optimization.
Non-stoichiometric perovskite oxides have been studied as a new family of redox oxides for solar thermochemical hydrogen (STCH) production owing to their favourable thermodynamic properties. However, conventional perovskite oxides suffer from limited phase stability and kinetic properties, and poor cyclability. Here, we report a strategy of introducing A-site multi-principal-component mixing to develop a high-entropy perovskite oxide, (La1/6Pr1/6Nd1/6Gd1/6Sr1/6Ba1/6)MnO3 (LPNGSB_Mn), which shows desirable thermodynamic and kinetics properties as well as excellent phase stability and cycling durability. LPNGSB_Mn exhibits enhanced hydrogen production (?77.5 mmol moloxide?1) compared to (La2/3Sr1/3)MnO3 (?53.5 mmol moloxide?1) in a short 1 hour redox duration and high STCH and phase stability for 50 cycles. LPNGSB_Mn possesses a moderate enthalpy of reduction (252.51-296.32 kJ (mol O)?1), a high entropy of reduction (126.95-168.85 J (mol O)?1 K?1), and fast surface oxygen exchange kinetics. All A-site cations do not show observable valence changes during the reduction and oxidation processes. This research preliminarily explores the use of one A-site high-entropy perovskite oxide for STCH.
We present a graph neural network approach that fully automates the prediction of defect formation enthalpies for any crystallographic site from the ideal crystal structure, without the need to create defected atomic structure models as input. Here we used density functional theory reference data for vacancy defects in oxides, to train a defect graph neural network (dGNN) model that replaces the density functional theory supercell relaxations otherwise required for each symmetrically unique crystal site. Interfaced with thermodynamic calculations of reduction entropies and associated free energies, the dGNN model is applied to the screening of oxides in the Materials Project database, connecting the zero-kelvin defect enthalpies to high-temperature process conditions relevant for solar thermochemical hydrogen production and other energy applications. The dGNN approach is applicable to arbitrary structures with an accuracy limited principally by the amount and diversity of the training data, and it is generalizable to other defect types and advanced graph convolution architectures. It will help to tackle future materials discovery problems in clean energy and beyond.
Wexler, Robert B.; Sai Gautam, Gopalakrishnan; Bell, Robert T.; Shulda, Sarah; Strange, Nicholas A.; Trindell, Jamie T.; Sugar, Joshua D.; Nygren, Eli; Sainio, Sami; Mcdaniel, Anthony H.; Ginley, David; Carter, Emily A.; Stechel, Ellen B.
Modeling-driven design of redox-active off-stoichiometric oxides for solar thermochemical H2 production (STCH) seldom has resulted in empirical demonstration of competitive materials. We report the theoretical prediction and experimental evidence that the perovskite Ca2/3Ce1/3Ti1/3Mn2/3O3 is synthesizable with high phase purity, stable, and has desirable redox thermodynamics for STCH, with a predicted average neutral oxygen vacancy (VO) formation energy, Ev = 3.30 eV. Flow reactor experiments suggest potentially comparable or greater H2 production capacity than recent promising Sr-La-Mn-Al and Ba-Ce-Mn metal oxide perovskites. Utilizing quantum-based modeling of a solid solution on both A and B sub-lattices, we predict the impact of nearest-neighbor composition on Ev and determine that A-site Ce4+ reduction dominates the redox-activity of Ca2/3Ce1/3Ti1/3Mn2/3O3. X-ray absorption spectroscopy measurements provide evidence that supports these predictions and reversible Ce4+-to-Ce3+ reduction. Our models predict that Ce4+ reduces even when it is not nearest-neighbor to the VO, suggesting that refinement of Ce stoichiometry has the possibility of further enhancing performance.
In this report, we investigate the thermal reduction of the octahedral perovskite BaCe0.25Mn0.75O3(BCM) using in situ electron energy loss spectroscopy (EELS) in an aberration-corrected transmission electron microscope (TEM). The 12R-polytype of BCM is known to demonstrate high solar thermochemical hydrogen production capacity. In situ EELS measurements show that Mn is the active redox cation in BCM, undergoing thermal reduction from Mn4+to Mn3+during heating to 700 °C inside the TEM under a high vacuum. The progressive reduction of Mn4+during oxygen vacancy (Ov) formation was monitored as a function of temperature. Additionally, atomic-resolution scanning transmission electron microscopy identified two different types of twin boundaries present in the oxidized and reduced form of 12R-BCM, respectively. These two types of twin boundaries were shown, via computational modeling, to modulate the site-specific Ovformation energies in 12R-BCM. It is concluded that these types of atomic defects provide sites more energetically favorable for Ovformation during thermal reduction.
Sandia National Laboratories in collaboration with the National Renewable Energy Laboratory outline a framework for developing a solar fuels roadmap based on novel concepts for hybridizing gas-splitting thermochemical cycle s with high-temperature electro chemical steps. We call this concept SoHyTEC, a Solar Hybrid Thermochemical-Electrochemical Cycle. The strategy focuses on transforming purely thermochemical cycles that split water (H2O) and carbon dioxide (CO2) to produce hydrogen (H 2 ) and carbon monoxide (CO) , respectively, the fundamental chemical building blocks for diverse fuels and chemicals , by substituting thermochemical reactions with high-temperature electrochemical steps. By invoking high-temperature electrochemistry, the energy required to complete the gas-splitting cycle is divided into a thermal component (process temperature) and an electrical component (applied voltage). These components, sourced from solar energy, are independently variable knobs to maximize overall process efficiency. Furthermore, a small applied voltage can reduce cycle process temperature by hundreds of degrees , opening the door to cost-effective solar concentrators and practical receiver/reactor de signs. Using the SoHyTEC concept as a backdrop, we outline a framework that advocates developing methods for automating information gathering, critically evaluating thermochemical cycles for adapting into SoHyTEC, establishing requirements based on thermodynamic analysis, and developing a model-based approach to benchmarking a SoHyTEC system against a baseline concentrating solar thermal integrated electrolysis plant. We feel these framework elements are a necessary precursor to creating a robust and adaptive technology development roadmap for producing solar fuels using SoHyTEC. In one example, we introduce high-temperature electrochemistry as a method to manipulate a fully stoichiometric two-step metal oxide cycle that circumvents costly separation processes and ultra-high cycle temperatures. We also identify and group water-splitting chemistries that are conceptually amenable to hybridization.
